Brain tumors have become the most deadly form of cancer in childhood, with its mortality currently exceeding that of leukemia. In addition, the malignant gliomas, the most frequent brain tumor in adults are associated with a very high mortality rate?only 2% of patients survive three years, which is one of the poorest prognoses across diseases. A silver lining can be seen with recent progress made in the treatment of leukemia, through usage of new drugs designed to specifically target neoplastic cells. In the case of brain cancer, the blood brain barrier (BBB) has been a vexing obstacle to the delivery of therapeutic agents, and thus, it is extremely difficult to achieve an adequate concentration of drug in brain tumors even if there is molecular targeting to tumor cells. Because of this, it is of the utmost importance to find a way to deliver adequate/high doses of therapeutics to brain tumors. While the intra-arterial therapy of gliomas after hyperosmotic blood-brain barrier (BBB) opening was attempted for years, the variability of results prevented widespread adoption of this technique, but of note is that at that time only non-selective infusions have been available. Recent advances in catheter design allows for super-selective, very distal catheter placement within the cerebral vasculature opening up opportunity for the resurrection of this method as a strategy to combat gliomas. It has been recently shown that single intra- arterial infusion of therapeutic agent matched the progression-free-survival (PFS) of traditional intravenous chemotherapy at a fraction of systemic adverse effects. However, until now the intra-arterial infusions of therapeutic agents are performed under X-ray fluoroscopy guidance, but the low sensitivity of X-ray contrast agents allows only detection of relatively large vessels, while the high sensitivity of MR contrast agents enables the direct visualization of the brain parenchyma supplied by the catheter, therefore easy overlay over the volume of the tumor. We have previously shown in a rabbit that MRI guidance is instrumental in highlighting the brain territory supplied by the catheter, which match the subsequent BBB opening area, which altogether makes the procedure predictable and precise. Here we propose to investigate through head-to-head comparison of the current clinically used X-ray based protocol with our MRI guidance- based approach whether its implementation for treatment of glioma will improve therapeutic outcome. Indeed, our preliminary results revealed that perfusion territory is variable and the target can be easily missed if no MRI guidance is provided and that may result in both suboptimal treatment and adverse effects. We will also optimize the intra-arterial dose of therapeutic agent based on the ratio of tumor shrinkage to injury of peritumoral healthy tissue. The ultimate goal of the study is to develop a clinic-ready protocol to replace the present X-ray based technique in the current clinical trial (NCT01688401) for intra-arterial treatment led by dr Pearl, a co-investigator on this proposal, which gives a promise for a very short path from bench to bed.